LiPO: Listwise Preference Optimization through Learning-to-Rank.
Tianqi LiuZhen QinJunru WuJiaming ShenMisha KhalmanRishabh JoshiYao ZhaoMohammad SalehSimon BaumgartnerJialu LiuPeter J. LiuXuanhui WangPublished in: CoRR (2024)
Keyphrases
- learning to rank
- direct optimization
- directly optimize
- ranking functions
- loss function
- information retrieval
- evaluation measures
- ranking svm
- document retrieval
- evaluation metrics
- learning to rank algorithms
- relevance judgments
- supervised learning
- query dependent
- balancing exploration and exploitation
- user feedback
- collaborative filtering
- ranking algorithm
- information retrieval systems
- digital libraries
- search engine